Description

This Zenodo repository contains the complete dataset, pre-trained models, experimental results, and manuscript figures required to reproduce the findings presented in our paper: "Unveiling the Potential of End-to-End Ocean Wave Modeling via a Hybrid Framework".

This single archive provides all the necessary components to be used with our NeuralWave Mini code, which is publicly available on GitHub. The code repository contains all the Python scripts for modeling, inference, and figure generation.

Main Code Repository: https://github.com/GaryYang77/NeuralWave-Mini.git

Quick Start Guide

1.  Clone the Code Repository from GitHub:

git clone https://github.com/GaryYang77/NeuralWave-Mini.git
cd NeuralWave-Mini

2.  Download this archive from the current Zenodo page.

3.  Extract and Place all Folders into the root of the cloned NeuralWave_mini directory. Your final project structure should look like this:

NeuralWave-Mini/
├── datasets/                    # <--- From this Zenodo download
├── results/                     # <--- From this Zenodo download
├── paper_figures/               # <--- From this Zenodo download
├── welltrained_case_models/     # <--- From this Zenodo download
├── models/                      # (From GitHub)
├── figure_generation/           # (From GitHub)
└── ... (other code files)

4. Follow the Instructions in the GitHub repository's README.md file to set up the environment and run the experiments.

Contents of this Archive

This archive (.zip or .tar.gz) contains the following directories:

• /datasets/: The complete training, validation, and test datasets.
• /results/: The full set of model prediction results in NetCDF (.nc) format.
• /paper_figures/: The final figures as they appear in the manuscript.
• /welltrained_case_models/: All pre-trained model weights for NeuralWave, EarthFormer, and U-Net.